@deeprelay/sdk 0.4.1 → 0.4.3

package/.openapi-generator/FILES

@@ -8,6 +8,7 @@ docs/AlertsApi.md
 docs/ApiKey.md
 docs/ApiKeyCreate.md
 docs/ApiKeyCreated.md
+docs/ApiKeyFallbackEntry.md
 docs/ApiKeyList.md
 docs/ApiKeyScopes.md
 docs/ApiKeyUpdate.md
@@ -51,6 +52,7 @@ docs/PeakWindow.md
 docs/Problem.md
 docs/RefereeStatus.md
 docs/Referral.md
+docs/RequestAttempt.md
 docs/RequestLogEntry.md
 docs/RequestLogPage.md
 docs/RequestsApi.md
@@ -107,6 +109,7 @@ src/models/AlertSettingsUpdate.ts
 src/models/ApiKey.ts
 src/models/ApiKeyCreate.ts
 src/models/ApiKeyCreated.ts
+src/models/ApiKeyFallbackEntry.ts
 src/models/ApiKeyList.ts
 src/models/ApiKeyScopes.ts
 src/models/ApiKeyUpdate.ts
@@ -146,6 +149,7 @@ src/models/PeakWindow.ts
 src/models/Problem.ts
 src/models/RefereeStatus.ts
 src/models/Referral.ts
+src/models/RequestAttempt.ts
 src/models/RequestLogEntry.ts
 src/models/RequestLogPage.ts
 src/models/ResponseFormatJSONObject.ts

package/README.md

@@ -1,4 +1,4 @@
-# @deeprelay/sdk@0.4.1
+# @deeprelay/sdk@0.4.3
 
 A TypeScript SDK client for the api.deeprelay.ai API.
 
@@ -99,6 +99,7 @@ All URIs are relative to *https://api.deeprelay.ai/v1*
 - [ApiKey](docs/ApiKey.md)
 - [ApiKeyCreate](docs/ApiKeyCreate.md)
 - [ApiKeyCreated](docs/ApiKeyCreated.md)
+- [ApiKeyFallbackEntry](docs/ApiKeyFallbackEntry.md)
 - [ApiKeyList](docs/ApiKeyList.md)
 - [ApiKeyScopes](docs/ApiKeyScopes.md)
 - [ApiKeyUpdate](docs/ApiKeyUpdate.md)
@@ -138,6 +139,7 @@ All URIs are relative to *https://api.deeprelay.ai/v1*
 - [Problem](docs/Problem.md)
 - [RefereeStatus](docs/RefereeStatus.md)
 - [Referral](docs/Referral.md)
+- [RequestAttempt](docs/RequestAttempt.md)
 - [RequestLogEntry](docs/RequestLogEntry.md)
 - [RequestLogPage](docs/RequestLogPage.md)
 - [ResponseFormatJSONObject](docs/ResponseFormatJSONObject.md)
@@ -192,7 +194,7 @@ and is automatically generated by the
 [OpenAPI Generator](https://openapi-generator.tech) project:
 
 - API version: `1.0.0`
-- Package version: `0.4.1`
+- Package version: `0.4.3`
 - Generator version: `7.24.0`
 - Build package: `org.openapitools.codegen.languages.TypeScriptFetchClientCodegen`
 

package/dist/apis/ApiKeysApi.d.ts

@@ -95,12 +95,12 @@ export declare class ApiKeysApi extends runtime.BaseAPI {
      */
     updateApiKeyRequestOpts(requestParameters: UpdateApiKeyRequest): Promise<runtime.RequestOpts>;
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKeyRaw(requestParameters: UpdateApiKeyRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<ApiKey>>;
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKey(requestParameters: UpdateApiKeyRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<ApiKey>;

package/dist/apis/ApiKeysApi.js

@@ -260,7 +260,7 @@ class ApiKeysApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKeyRaw(requestParameters, initOverrides) {
@@ -271,7 +271,7 @@ class ApiKeysApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKey(requestParameters, initOverrides) {

package/dist/apis/BillingApi.d.ts

@@ -134,12 +134,12 @@ export declare class BillingApi extends runtime.BaseAPI {
      */
     getReferralRequestOpts(): Promise<runtime.RequestOpts>;
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferralRaw(initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<Referral>>;
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferral(initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<Referral>;

package/dist/apis/BillingApi.js

@@ -341,7 +341,7 @@ class BillingApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferralRaw(initOverrides) {
@@ -352,7 +352,7 @@ class BillingApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferral(initOverrides) {

package/dist/apis/InferenceApi.d.ts

@@ -28,6 +28,7 @@ export interface CancelVideoRequest {
 export interface CreateChatCompletionRequest {
     chatCompletionRequest: ChatCompletionRequest;
     xClientRequestId?: string;
+    xDeeprelayFallbackModels?: string;
 }
 export interface CreateEmbeddingsRequest {
     embeddingsRequest: EmbeddingsRequest;
@@ -84,12 +85,12 @@ export declare class InferenceApi extends runtime.BaseAPI {
      */
     createChatCompletionRequestOpts(requestParameters: CreateChatCompletionRequest): Promise<runtime.RequestOpts>;
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletionRaw(requestParameters: CreateChatCompletionRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<ChatCompletionResponse>>;
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletion(requestParameters: CreateChatCompletionRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<ChatCompletionResponse>;

package/dist/apis/InferenceApi.js

@@ -100,6 +100,9 @@ class InferenceApi extends runtime.BaseAPI {
             if (requestParameters['xClientRequestId'] != null) {
                 headerParameters['X-Client-Request-Id'] = String(requestParameters['xClientRequestId']);
             }
+            if (requestParameters['xDeeprelayFallbackModels'] != null) {
+                headerParameters['X-Deeprelay-Fallback-Models'] = String(requestParameters['xDeeprelayFallbackModels']);
+            }
             if (this.configuration && this.configuration.accessToken) {
                 const token = this.configuration.accessToken;
                 const tokenString = yield token("bearerAuth", []);
@@ -118,7 +121,7 @@ class InferenceApi extends runtime.BaseAPI {
         });
     }
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletionRaw(requestParameters, initOverrides) {
@@ -129,7 +132,7 @@ class InferenceApi extends runtime.BaseAPI {
         });
     }
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletion(requestParameters, initOverrides) {

package/dist/apis/RequestsApi.d.ts

@@ -21,6 +21,7 @@ export interface ExportRequestsRequest {
     clientRequestId?: string;
     from?: Date;
     to?: Date;
+    fallback?: boolean;
 }
 export interface GetRequestRequest {
     id: string;
@@ -34,6 +35,7 @@ export interface ListRequestsRequest {
     clientRequestId?: string;
     from?: Date;
     to?: Date;
+    fallback?: boolean;
     cursor?: string;
     limit?: number;
 }
@@ -46,12 +48,12 @@ export declare class RequestsApi extends runtime.BaseAPI {
      */
     exportRequestsRequestOpts(requestParameters: ExportRequestsRequest): Promise<runtime.RequestOpts>;
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequestsRaw(requestParameters: ExportRequestsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<string>>;
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequests(requestParameters?: ExportRequestsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<string>;

package/dist/apis/RequestsApi.js

@@ -60,6 +60,9 @@ class RequestsApi extends runtime.BaseAPI {
             if (requestParameters['to'] != null) {
                 queryParameters['to'] = requestParameters['to'].toISOString();
             }
+            if (requestParameters['fallback'] != null) {
+                queryParameters['fallback'] = requestParameters['fallback'];
+            }
             const headerParameters = {};
             if (this.configuration && this.configuration.accessToken) {
                 const token = this.configuration.accessToken;
@@ -78,7 +81,7 @@ class RequestsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequestsRaw(requestParameters, initOverrides) {
@@ -94,7 +97,7 @@ class RequestsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequests() {
@@ -181,6 +184,9 @@ class RequestsApi extends runtime.BaseAPI {
             if (requestParameters['to'] != null) {
                 queryParameters['to'] = requestParameters['to'].toISOString();
             }
+            if (requestParameters['fallback'] != null) {
+                queryParameters['fallback'] = requestParameters['fallback'];
+            }
             if (requestParameters['cursor'] != null) {
                 queryParameters['cursor'] = requestParameters['cursor'];
             }

package/dist/esm/apis/ApiKeysApi.d.ts

@@ -95,12 +95,12 @@ export declare class ApiKeysApi extends runtime.BaseAPI {
      */
     updateApiKeyRequestOpts(requestParameters: UpdateApiKeyRequest): Promise<runtime.RequestOpts>;
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKeyRaw(requestParameters: UpdateApiKeyRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<ApiKey>>;
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKey(requestParameters: UpdateApiKeyRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<ApiKey>;

package/dist/esm/apis/ApiKeysApi.js

@@ -257,7 +257,7 @@ export class ApiKeysApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKeyRaw(requestParameters, initOverrides) {
@@ -268,7 +268,7 @@ export class ApiKeysApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Changes a key\'s name, budget, model allowlist, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
+     * Changes a key\'s name, budget, model allowlist, backup models, rate limit or expiry. An absent field is left unchanged; an explicit `null` clears `budget_cents`, `allowed_models`, `fallback_models` or `expires_at`. `scopes` cannot be changed after minting, and unknown fields are refused (422), so a typo never looks like a successful change. Requires the `keys:manage` capability (403 `keys_manage_required` otherwise); a member can only change keys they created.
      * Update an API key\'s name or limits
      */
     updateApiKey(requestParameters, initOverrides) {

package/dist/esm/apis/BillingApi.d.ts

@@ -134,12 +134,12 @@ export declare class BillingApi extends runtime.BaseAPI {
      */
     getReferralRequestOpts(): Promise<runtime.RequestOpts>;
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferralRaw(initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<Referral>>;
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferral(initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<Referral>;

package/dist/esm/apis/BillingApi.js

@@ -338,7 +338,7 @@ export class BillingApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferralRaw(initOverrides) {
@@ -349,7 +349,7 @@ export class BillingApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; when a friend signs up through it and spends `qualify_spend_cents` on inference, the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. The referrer also earns `commission_pct` percent of every real deposit the friend makes, credited as spend-only promotional credit when the deposit clears (`commission_earned_cents`). Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
+     * Returns the organization\'s shareable referral code and invite link (minting the code on first read), the program terms, the referrer\'s stats, and — when this organization itself signed up through someone else\'s link — its own progress toward the referee reward. Requires the `billing:read` scope. The organization is taken from the authenticated API key, never from a parameter.  The program: share the invite link; once a friend who signed up through it has made `qualify_deposit_cents` of net real deposits (card, auto-pay or crypto, less refunds), the referrer receives `reward_cents` and the friend receives `referee_reward_cents`, both as non-withdrawable credit, after a `hold_days` chargeback hold. Both rewards are reversed if the qualifying deposit is later refunded. The deposit commission is retired and `commission_pct` is 0; when it is non-zero the referrer also earns that percent of every real deposit the friend makes. `commission_earned_cents` still reports commission earned before retirement. Program rewards never apply to the subscription. The terms ride on the wire so a client never hard-codes the amounts.  `referred` is ABSENT (not null) for an organization nobody referred — key on its presence. This is the same contract the dashboard\'s referral card reads.
      * Get the org referral code, invite link, terms and stats
      */
     getReferral(initOverrides) {

package/dist/esm/apis/InferenceApi.d.ts

@@ -28,6 +28,7 @@ export interface CancelVideoRequest {
 export interface CreateChatCompletionRequest {
     chatCompletionRequest: ChatCompletionRequest;
     xClientRequestId?: string;
+    xDeeprelayFallbackModels?: string;
 }
 export interface CreateEmbeddingsRequest {
     embeddingsRequest: EmbeddingsRequest;
@@ -84,12 +85,12 @@ export declare class InferenceApi extends runtime.BaseAPI {
      */
     createChatCompletionRequestOpts(requestParameters: CreateChatCompletionRequest): Promise<runtime.RequestOpts>;
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletionRaw(requestParameters: CreateChatCompletionRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<ChatCompletionResponse>>;
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletion(requestParameters: CreateChatCompletionRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<ChatCompletionResponse>;

package/dist/esm/apis/InferenceApi.js

@@ -97,6 +97,9 @@ export class InferenceApi extends runtime.BaseAPI {
             if (requestParameters['xClientRequestId'] != null) {
                 headerParameters['X-Client-Request-Id'] = String(requestParameters['xClientRequestId']);
             }
+            if (requestParameters['xDeeprelayFallbackModels'] != null) {
+                headerParameters['X-Deeprelay-Fallback-Models'] = String(requestParameters['xDeeprelayFallbackModels']);
+            }
             if (this.configuration && this.configuration.accessToken) {
                 const token = this.configuration.accessToken;
                 const tokenString = yield token("bearerAuth", []);
@@ -115,7 +118,7 @@ export class InferenceApi extends runtime.BaseAPI {
         });
     }
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletionRaw(requestParameters, initOverrides) {
@@ -126,7 +129,7 @@ export class InferenceApi extends runtime.BaseAPI {
         });
     }
     /**
-     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.
+     * OpenAI-compatible chat completion. Parameters fall into three groups:  - **Semantic parameters are gated per model.** `response_format`,   `tools`/`tool_choice`, `logprobs`/`top_logprobs` and `image_url`   content parts are sent only to a model whose `/v1/models` entry   lists the capability in `supported_parameters` (`vision` for   images). A model without it returns 400 `unsupported_parameter`   whose message names the models that do support it. When the model   supports the parameter but none of its currently routable upstreams   does, the request fails with 503 `upstream_unavailable` rather than   being silently degraded. - **Hints are forwarded or dropped, never rejected.**   `reasoning_effort`, `presence_penalty`, `frequency_penalty`, `seed`   and `parallel_tool_calls` are forwarded to an upstream that accepts   them and otherwise dropped. Every drop is reported in the   `x-deeprelay-ignored-params` response header. - **Rejected everywhere.** `logit_bias` and `function_call` return   400 `unsupported_parameter`; `input_audio` and `file` content parts   return 400 `unsupported_content_part`.  A non-strict `json_schema` sent to a model that can only guarantee JSON is served as `json_object` and reported as `response_format:downgraded`; a strict one is rejected. A `json_schema` over 64 KiB or nested deeper than 32 levels returns 400 `unsupported_parameter` (param `response_format`). An upstream rejection of a request the relay accepted returns 400 `upstream_rejected_request`. A non-chat model id returns 404 `model_not_found`. The request body is capped at 32 MiB (413 `request_too_large`).  Every successful response carries `usage.cost` and `usage.cost_details` (on a stream, only in the final usage chunk). See https://deeprelay.ai/docs/api/inference#openai-compatibility.  **Backup models.** When the API key has `fallback_models`, or the request sends `models` or X-Deeprelay-Fallback-Models, and every provider for the requested model fails before the first token with a temporary error (unavailable, timed out, rate limited upstream, or stalled), deeprelay tries the next backup model. It never switches after the first byte, and never on a client error, a quota or budget refusal, or a content-filter result. A backup is billed at its own price under the same billing mode as the requested model. The body `model` and X-Deeprelay-Served-Model name the model that served; X-Deeprelay-Requested-Model names the one asked for, and X-Deeprelay-Fallback-Reason says why it was passed over. If no model serves, the response is the requested model\'s own error, with X-Deeprelay-Fallback-Attempted.
      * Create a chat completion (OpenAI-compatible)
      */
     createChatCompletion(requestParameters, initOverrides) {

package/dist/esm/apis/RequestsApi.d.ts

@@ -21,6 +21,7 @@ export interface ExportRequestsRequest {
     clientRequestId?: string;
     from?: Date;
     to?: Date;
+    fallback?: boolean;
 }
 export interface GetRequestRequest {
     id: string;
@@ -34,6 +35,7 @@ export interface ListRequestsRequest {
     clientRequestId?: string;
     from?: Date;
     to?: Date;
+    fallback?: boolean;
     cursor?: string;
     limit?: number;
 }
@@ -46,12 +48,12 @@ export declare class RequestsApi extends runtime.BaseAPI {
      */
     exportRequestsRequestOpts(requestParameters: ExportRequestsRequest): Promise<runtime.RequestOpts>;
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequestsRaw(requestParameters: ExportRequestsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<string>>;
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequests(requestParameters?: ExportRequestsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<string>;

package/dist/esm/apis/RequestsApi.js

@@ -57,6 +57,9 @@ export class RequestsApi extends runtime.BaseAPI {
             if (requestParameters['to'] != null) {
                 queryParameters['to'] = requestParameters['to'].toISOString();
             }
+            if (requestParameters['fallback'] != null) {
+                queryParameters['fallback'] = requestParameters['fallback'];
+            }
             const headerParameters = {};
             if (this.configuration && this.configuration.accessToken) {
                 const token = this.configuration.accessToken;
@@ -75,7 +78,7 @@ export class RequestsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequestsRaw(requestParameters, initOverrides) {
@@ -91,7 +94,7 @@ export class RequestsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
+     * Streams every request-log row matching the filters as a CSV attachment, oldest columns first: `request_id, client_request_id, occurred_at, surface, model, key_id, key_name, key_prefix, status, http_status, error_code, finish_reason, prompt_tokens, completion_tokens, cached_tokens, reasoning_tokens, quota_units, cost, plan_covered, ttft_ms, duration_ms, ignored_params, requested_model, served_model, fallback_reason`. Takes the same filters and visibility rules as `GET /requests`, without the cursor and limit.  The export is counted before anything is written. When the filters match more rows than the export cap, the answer is 422 `export_too_large` and the problem body carries `max_rows`; narrow the time range or add a filter, then export again.  Requires the `serverless:read` scope. Exports have their own, tighter per-key rate limit.
      * Export the request log as CSV
      */
     exportRequests() {
@@ -178,6 +181,9 @@ export class RequestsApi extends runtime.BaseAPI {
             if (requestParameters['to'] != null) {
                 queryParameters['to'] = requestParameters['to'].toISOString();
             }
+            if (requestParameters['fallback'] != null) {
+                queryParameters['fallback'] = requestParameters['fallback'];
+            }
             if (requestParameters['cursor'] != null) {
                 queryParameters['cursor'] = requestParameters['cursor'];
             }